Multidimensional penalized splines for survival models: illustration for net survival trend analyses.

BACKGROUND: In descriptive epidemiology, there are strong similarities between incidence and survival analyses. Because of the success of multidimensional penalized splines (MPSs) in incidence analysis, we propose in this pedagogical paper to show that MPSs are also very suitable for survival or net survival studies. METHODS: The use of MPSs is illustrated in cancer epidemiology in the context of survival trends studies that require specific statistical modelling. We focus on two examples (cervical and colon cancers) using survival data from the French cancer registries (cases 1990-2015). The dynamic of the excess mortality hazard according to time since diagnosis was modelled using an MPS of time since diagnosis, age at diagnosis and year of diagnosis. Multidimensional splines bring the flexibility necessary to capture any trend patterns while penalization ensures selecting only the complexities necessary to describe the data. RESULTS: For cervical cancer, the dynamic of the excess mortality hazard changed with the year of diagnosis in opposite ways according to age: this led to a net survival that improved in young women and worsened in older women. For colon cancer, regardless of age, excess mortality decreases with the year of diagnosis but this only concerns mortality at the start of follow-up. CONCLUSIONS: MPSs make it possible to describe the dynamic of the mortality hazard and how this dynamic changes with the year of diagnosis, or more generally with any covariates of interest: this gives essential epidemiological insights for interpreting results. We use the R package survPen to do this type of analysis.

Monitoring multidimensional aspects of quality of life after cancer immunotherapy: protocol for the international multicentre, observational QUALITOP cohort study.

INTRODUCTION: Immunotherapies, such as immune checkpoint inhibitors and chimeric antigen receptor T-cell therapy, have significantly improved the clinical outcomes of various malignancies. However, they also cause immune-related adverse events (irAEs) that can be challenging to predict, prevent and treat. Although they likely interact with health-related quality of life (HRQoL), most existing evidence on this topic has come from clinical trials with eligibility criteria that may not accurately reflect real-world settings. The QUALITOP project will study HRQoL in relation to irAEs and its determinants in a real-world study of patients treated with immunotherapy. METHODS AND ANALYSIS: This international, observational, multicentre study takes place in France, the Netherlands, Portugal and Spain. We aim to include about 1800 adult patients with cancer treated with immunotherapy in a specifically recruited prospective cohort, and to additionally obtain data from historical real-world databases (ie, databiobanks) and medical administrative registries (ie, national cancer registries) in which relevant data regarding other adult patients with cancer treated with immunotherapy has already been stored. In the prospective cohort, clinical health status, HRQoL and psychosocial well-being will be monitored until 18 months after treatment initiation through questionnaires (at baseline and 3, 6, 12 and 18 months thereafter), and by data extraction from electronic patient files. Using advanced statistical methods, including causal inference methods, artificial intelligence algorithms and simulation modelling, we will use data from the QUALITOP cohort to improve the understanding of the complex relationships among treatment regimens, patient characteristics, irAEs and HRQoL. ETHICS AND DISSEMINATION: All aspects of the QUALITOP project will be conducted in accordance with the Declaration of Helsinki and with ethical approval from a suitable local ethics committee, and all patients will provide signed informed consent. In addition to standard dissemination efforts in the scientific literature, the data and outcomes will contribute to a smart digital platform and medical data lake. These will (1) help increase knowledge about the impact of immunotherapy, (2) facilitate improved interactions between patients, clinicians and the general population and (3) contribute to personalised medicine. TRIAL REGISTRATION NUMBER: NCT05626764.

Is the Social Gradient in Net Survival Observed in France the Result of Inequalities in Cancer-Specific Mortality or Inequalities in General Mortality?

BACKGROUND: In cancer net survival analyses, if life tables (LT) are not stratified based on socio-demographic characteristics, then the social gradient in mortality in the general population is ignored. Consequently, the social gradient estimated on cancer-related excess mortality might be inaccurate. We aimed to evaluate whether the social gradient in cancer net survival observed in France could be attributable to inaccurate LT. METHODS: Deprivation-specific LT were simulated, applying the social gradient in the background mortality due to external sources to the original French LT. Cancer registries' data from a previous French study were re-analyzed using the simulated LT. Deprivation was assessed according to the European Deprivation Index (EDI). Net survival was estimated by the Pohar-Perme method and flexible excess mortality hazard models by using multidimensional penalized splines. RESULTS: A reduction in net survival among patients living in the most-deprived areas was attenuated with simulated LT, but trends in the social gradient remained, except for prostate cancer, for which the social gradient reversed. Flexible modelling additionally showed a loss of effect of EDI upon the excess mortality hazard of esophagus, bladder and kidney cancers in men and bladder cancer in women using simulated LT. CONCLUSIONS: For most cancers the results were similar using simulated LT. However, inconsistent results, particularly for prostate cancer, highlight the need for deprivation-specific LT in order to produce accurate results.

Mortality and Associated Causes in Hemophagocytic Lymphohistiocytosis: A Multiple-Cause-of-Death Analysis in France.

Hemophagocytic lymphohistiocytosis (HLH) is a severe hyperinflammatory syndrome with an overall mortality rate of 40%. A multiple-cause-of-death analysis allows for the characterization of mortality and associated causes over an extended period. Death certificates, collected between 2000 and 2016 by the French Epidemiological Centre for the Medical Causes of Death (CepiDC, Inserm), containing the ICD10 codes for HLH (D76.1/2), were used to calculate HLH-related mortality rates and to compare them with the general population (observed/expected ratios, O/E). HLH was mentioned in 2072 death certificates as the underlying cause of death (UCD, n = 232) or as a non-underlying cause of death (NUCD, n = 1840). The mean age at death was 62.4 years. The age-standardized mortality rate was 1.93/million person-years and increased over the study period. When HLH was an NUCD, the most frequently associated UCDs were hematological diseases (42%), infections (39.4%), and solid tumors (10.4%). As compared to the general population, HLH decedents were more likely to have associated CMV infections or hematological diseases. The increase in mean age at death over the study period indicates progress in diagnostic and therapeutic management. This study suggests that the prognosis of HLH may be at least partially related to coexisting infections and hematological malignancies (either as causes of HLH or as complications).

Clinical characteristics and outcome of 318 families with familial monoclonal gammopathy: A multicenter Intergroupe Francophone du Myélome study.

Familial forms of monoclonal gammopathy, defined as multiple myeloma (MM) or Monoclonal Gammopathy of Undetermined Significance (MGUS), are relatively infrequent and most series reported in the literature describe a limited number of families. MM rarely occurs in a familial context. MGUS is observed much more commonly, which can in some cases evolve toward full-blown MM. Although recurrent cytogenetic abnormalities have been described in tumor cells of sporadic cases of MM, the pathogenesis of familial MM remains largely unexplained. In order to identify genetic factors predisposing to familial monoclonal gammopathy, the Intergroupe Francophone du Myélome identified 318 families with at least two confirmed cases of monoclonal gammopathy. There were 169 families with parent/child pairs and 164 families with cases in at least two siblings, compatible with an autosomal transmission. These familial cases were compared with sporadic cases who were matched for age at diagnosis, sex and immunoglobulin isotype, with 10 sporadic cases for each familial case. The gender distribution, age and immunoglobulin subtypes of familial cases were unremarkable in comparison to sporadic cases. With a median follow-up of 7.4 years after diagnosis, the percentage of MGUS cases having evolved to MM was 3%. The median overall survival of the 148 familial MM cases was longer than that of matched sporadic cases, with projected values of 7.6 and 16.1 years in patients older and younger than 65 years, respectively. These data suggest that familial cases of monoclonal gammopathy are similar to sporadic cases in terms of clinical presentation and carry a better prognosis.

How do age and social environment affect the dynamics of death hazard and survival in patients with breast or gynecological cancer in France?

Several studies have investigated the association between net survival (NS) and social inequalities in people with cancer, highlighting a varying influence of deprivation depending on the type of cancer studied. However, few of these studies have accounted for the effect of social inequalities over the follow-up period, and/or according to the age of the patients. Thus, using recent and more relevant statistical models, we investigated the effect of social environment on NS in women with breast or gynecological cancer in France. The data were derived from population-based cancer registries, and women diagnosed with breast or gynecological cancer between 2006 and 2009 were included. We used the European deprivation index (EDI), an aggregated index, to define the social environment of the women included. Multidimensional penalized splines were used to model excess mortality hazard. We observed a significant effect of the EDI on NS in women with breast cancer throughout the follow-up period, and especially at 1.5 years of follow-up in women with cervical cancer. Regarding corpus uteri and ovarian cancer patients, the effect of deprivation on NS was less pronounced. These results highlight the impact of social environment on NS in women with breast or gynecological cancer in France thanks to a relevant statistical approach, and identify the follow-up periods during which the social environment may have a particular influence. These findings could help investigate targeted actions for each cancer type, particularly in the most deprived areas, at the time of diagnosis and during follow-up.

Outcomes and survival of a modern dual mobility cup and uncemented collared stem in displaced femoral neck fractures at a minimum 5-year follow-up.

INTRODUCTION: The choice of implant type for total hip replacement in the treatment of femoral neck fractures remains debated. Some authors advocate for the systematic use of cemented stems, while others do not use dual mobility first-line. We therefore conducted a retrospective study using a dual mobility cup (DMC) and an uncemented collared stem (UCS) in order to: (1) confirm the low dislocation rate in this indication, (2) assess other surgical complications, in particular periprosthetic fractures, (3) ensure that these benefits are maintained over time, at a minimum follow-up of 5 years and, (4) assess the rate of revision of the implants. HYPOTHESIS: Our hypothesis was that the dual mobility dislocation rate for the treatment of femoral neck fractures was lower than for bipolar hemiarthroplasties or single mobility hip prostheses. PATIENTS AND METHODS: A retrospective study of 244 femoral neck fractures (242 patients) treated with DMC and UCS was conducted, between 2013 and 2014. The mean age was 83±10 years (60-104). The occurrence of dislocation, periprosthetic fracture, infection of the surgical site, loosening, reoperation and revision were investigated. The HOOS Joint Replacement (JR) score was collected. The cumulative incidence with mortality was used as a competing risk. RESULTS: The mean follow-up was 6 years±0.5 (5-7). At the last follow-up, 108 patients (50%) had died. Twenty-three patients (9.5%) were lost to follow-up. One case of symptomatic aseptic loosening of DMC was observed. The cumulative incidence of dislocations and periprosthetic fractures at 5 years were 2% (95% CI: 0.9-5.4) and 3% (95% CI: 1.2-6), respectively. The 5-year cumulative incidence of surgical site infections was 3.5% (95% CI: 1.8-7). The cumulative incidence of reoperations at 5 years was 7% (95% CI: 4.5-11). The causes of reoperation were periprosthetic fracture (n=6), infection (n=8), postoperative hematoma (n=2) and cup malposition (n=2). The cumulative incidence of a revision at 5 years was 2.7% (95% CI: 1.2-6). The cumulative incidence of a surgical complication from any cause at 5 years was 9% (95% CI: 6.7-14.8). The mean HOOS JR score was 79±5 (52-92). DISCUSSION: The cumulative incidence of dislocation at 5 years is low and other surgical complications (including periprosthetic fractures) do not increase during this period for DMC associated with UCS, in femoral neck fractures. The use of this type of implant is reliable in the treatment of femoral neck fractures. LEVEL OF EVIDENCE: IV; retrospective study without control group.

Socioeconomic Environment and Survival in Patients with Digestive Cancers: A French Population-Based Study.

Social inequalities are an important prognostic factor in cancer survival, but little is known regarding digestive cancers specifically. We aimed to provide in-depth analysis of the contextual social disparities in net survival of patients with digestive cancer in France, using population-based data and relevant modeling. Digestive cancers (n = 54,507) diagnosed between 2006-2009, collected through the French network of cancer registries, were included (end of follow-up 30 June 2013). Social environment was assessed by the European Deprivation Index. Multidimensional penalized splines were used to model excess mortality hazard. We found that net survival was significantly worse for individuals living in a more deprived environment as compared to those living in a less deprived one for esophageal, liver, pancreatic, colon and rectal cancers, and for stomach and bile duct cancers among females. Excess mortality hazard was up to 57% higher among females living in the most deprived areas (vs. least deprived) at 1 year of follow-up for bile duct cancer, and up to 21% higher among males living in the most deprived areas (vs. least deprived) regarding colon cancer. To conclude, we provide a better understanding of how the (contextual) social gradient in survival is constructed, offering new perspectives for tackling social inequalities in digestive cancer survival.

Association Between Short-, Intermediate-, and Long-term Mortality and Myocardial Injury After Noncardiac Surgery After Hip Fracture Surgery: A Retrospective Cohort.

BACKGROUND: For more than 20 years, hip fracture 1-year mortality has remained around 20%. An elevation of the postoperative troponin peak within 72 hours (myocardial injury after noncardiac surgery [MINS]) is associated with a greater risk of short-term mortality in the general population. However, there seem to be conflicting results in the specific population who undergo hip fracture surgery, with some studies finding an association between troponin and mortality and some not. The objective of the present study was to investigate the association of MINS and the short- (before 28th day), intermediate- (before 180th day), and long-term (before 365th day) mortality after hip fracture surgery. METHODS: We conducted a single-center retrospective cohort of patients undergoing hip fracture surgery from November 2013 to December 2015. MINS was defined as postoperative troponin peak within the 72 hours >5 ng/L. Four MINS subgroups were defined according to the value of troponin peak (ie, ≥5-<20, ≥20-<65, ≥65-<1000, and ≥1000 ng/L). To document the association between the different mortality terms and the troponin peak, odds ratio (OR) and adjusted OR (aOR) associated with their 95% confidence interval (CI) with the log of the scaled troponin peak within 72 hours were estimated, with and without patients presenting a postoperative acute coronary syndrome (ACS). Cox proportional hazards modeling was used to estimate hazard ratio (HR) and adjusted HR (aHR) of death between the no MINS and MINS subgroups. The adjustment was performed on the main confounding factors (ie, sex, American Society of Anesthesiologists [ASA] physical status, dementia status, age, and time from admission to surgery). RESULTS: Among 729 participants, the mean age was 83.1 (standard deviation [SD] = 10.8) years, and 77.4% were women; 30 patients presented an ACS (4%). Short-, intermediate-, and long-term mortality were at 5%, 16%, and 23%, respectively. The troponin peak was significantly associated with all terms of mortality before and after adjustment and before and after exclusion of patients presenting an ACS. HR and aHR for each subgroup of troponin level were significantly associated with an increased probability of survival, except for the 5 to 20 ng/L group for which aHR was not significant (1.75, 95% CI, 0.82-3.74). In the landmark analysis, there was still an association between survival at the 365th day and troponin peak after the short- and intermediate-term truncated mortality. CONCLUSIONS: MINS is associated with short-, intermediate-, and long-term mortality after hip fracture surgery. This could be a valuable indicator to determine the population at high risk of mortality that could benefit from targeted prevention and possible intervention.

Multidimensional penalized splines for incidence and mortality-trend analyses and validation of national cancer-incidence estimates.

BACKGROUND: Cancer-incidence and mortality-trend analyses require appropriate statistical modelling. In countries without a nationwide cancer registry, an additional issue is estimating national incidence from local-registry data. The objectives of this study were to (i) promote the use of multidimensional penalized splines (MPS) for trend analyses; (ii) estimate the national cancer-incidence trends, using MPS, from only local-registry data; and (iii) propose a validation process of these estimates. METHODS: We used an MPS model of age and year for trend analyses in France over 1990-2015 with a projection up to 2018. Validation was performed for 22 cancer sites and relied essentially on comparison with reference estimates that used the incidence/health-care ratio over the period 2011-2015. Alternative estimates that used the incidence/mortality ratio were also used to validate the trends. RESULTS: In the validation assessment, the relative differences of the incidence estimates (2011-2015) with the reference estimates were <5% except for testis cancer in men and < 7% except for larynx cancer in women. Trends could be correctly derived since 1990 despite incomplete histories in some registries. The proposed method was applied to estimate the incidence and mortality trends of female lung cancer and prostate cancer in France. CONCLUSIONS: The validation process confirmed the validity of the national French estimates; it may be applied in other countries to help in choosing the most appropriate national estimation method according to country-specific contexts. MPS form a powerful statistical tool for trend analyses; they allow trends to vary smoothly with age and are suitable for modelling simple as well as complex trends thanks to penalization. Detailed trend analyses of lung and prostate cancers illustrated the suitability of MPS and the epidemiological interest of such analyses.

Socioeconomic environment and disparities in cancer survival for 19 solid tumor sites: An analysis of the French Network of Cancer Registries (FRANCIM) data.

Social inequalities are concerning along the cancer continuum. In France, social gradient in health is particularly marked but little is known about social gradient in cancer survival. We aimed to investigate the influence of socioeconomic environment on cancer survival, for all cancers reported in the French Network of Cancer Registries. We analyzed 189,657 solid tumors diagnosed between 2006 and 2009, recorded in 18 registries. The European Deprivation Index (EDI), an ecological index measuring relative poverty in small geographic areas, assessed social environment. The EDI was categorized into quintiles of the national distribution. One- and five-year age-standardized net survival (ASNS) were estimated for each solid tumor site and deprivation quintile, among men and among women. We found that 5-year ASNS was lower among patients living in the most deprived areas compared to those living in the least deprived ones for 14/16 cancers among men and 16/18 cancers among women. The extent of cancer survival disparities according to deprivation varied substantially across the cancer sites. The reduction in ASNS between the least and the most deprived quintile reached 34% for liver cancer among men and 59% for bile duct cancer among women. For pancreas, stomach and esophagus cancer (among men), and ovary and stomach cancer (among women), deprivation gaps were larger at 1-year than 5-year survival. In conclusion, survival was worse in the most deprived areas for almost all cancers. Our results from population-based cancer registries data highlight the need for implementing actions to reduce social inequalities in cancer survival in France.